The amount of bone gained during childhood impacts greatly on lifetime skeletal health (Gruodyte-Raciene et al., 2013). Bone is a metabolically active tissue with continuous remodeling occurring throughout the lifespan (Jurimae, 2010). Bone mineral accrual increases substantially during the growing years peaking in velocity of accrual (peak bone mineral content velocity [PBMCV]) seven months after the attainment of peak height velocity (PHV) (Bailey, 1999). Bone mass then plateaus between the end of the second decade of life and the middle of the third; termed peak bone mass (PBM) (Bailey, 1999). Individuals who achieve a higher PBMCV during adolescence and a subsequently higher PBM have decreased fracture risk in later life (Weaver et al., 2016; Xu et al., 2016). Maximal increases in bone mineral accrual occur over a relatively brief period in the years surrounding PHV (Baxter-Jones et al., 2011; Jackowski et al., 2011a). Specifically, Baxter-Jones et al. (2011) showed that up to 40% of peak bone was accrued in a five year window surrounding attainment of PHV. Accordingly, puberty is an opportune time for bone strengthening (Jackowski et al., 2011a; Vaitkeviciute et al., 2014), and research should focus on strategies for maximizing peak bone mineral accrual during this growth period in order to lay the foundation for better adult bone health.
The accumulation of bone mineral during pubertal growth is influenced by a number of factors, including but not limited to: timing of pubertal maturation (Jackowski et al. 2011a), body composition (Ivuskans et al., 2013), nutritional status (Maimoun et al., 2014), endocrine function (Pomerants et al., 2007), habitual physical activity (Vaitkeviciute et al., 2016) and athletic training (Vosoberg et al., 2016). Therefore, some of these factors may influence bone mineral accrual negatively and some positively. The decrease in overall physical activity and increase in daily sedentary time has been shown to negatively influence bone mineral accrual in boys during puberty (Ivuskans et al., 2015; Vaitkeviciute et al., 2014), while mechanical loading of athletic training is a positive factor for skeletal strength and bone development during the same time period (Gruodyte et al., 2010a; Parm et al., 2011a; 2011b). In addition to the independent effect of mechanical loading on bone mass and density, increased mechanical loading as a result of athletic training may also influence structural changes in bone to increase bone strength in response to specific loading conditions (Weaver et al., 2016). Therefore, the effects of sport on bone health in children vary in relation to training modality, ranging from non-weight bearing (swimming) and low-impact activities (skiing) to high-impact activities (gymnastics) (Gruodyte et al., 2009; 2010b). Accordingly, regular high-impact weight-bearing athletic activity during pubertal growth plays an important role in maximizing bone mineral gain and may reduce the risk of osteoporosis in later life (Baxter-Jones et al., 2008; Erlandson et al., 2012a; 2012b). It appears that gymnastics training is especially osteogenic for bone development in children (Gruodyte-Racience et al., 2013), adolescents (Gruodyte et al., 2009) and adults (Soot et al., 2005), probably due to the high-volume, high-impact training and involvement at a relatively early age during growth (Tournis et al., 2010). However, intense athletic activity in growing and maturing gymnasts imposes several constraints such as training stress and maintenance of a relatively low fat mass (FM) in order to maximize gymnastics performance (Maimoun et al., 2014). Pubertal gymnasts are at risk of inadequate dietary intake, which in turn may have several consequences for endocrine function (Jurimae, 2014; Malina et al., 2013). Accordingly, a serious question is raised about the positive effects of regular gymnastics activities on overall health and especially normal bone mineral accrual during the growing years in age related elite, sub-elite and recreational gymnasts. It is not clear how prolonged high gymnastics activity during growth and maturation affects bone health later in adulthood.
To date, bone health in gymnasts has been measured by different imaging techniques. The assessment of bone health includes the two-dimensional measurement of a real bone mineral density (aBMD) and bone mineral content (BMC) measured by dual-energy X-ray absorptiometry (DXA) and/or the three-dimensional measurement of volumetric BMD (vBMD) and bone strength indices measured by peripheral quantitative computed tomography (pQCT) (Xu et al., 2016). DXA scans have also been used to assess the structural geometry of the proximal femur using the hip structural analysis program (Beck et al., 1990; Jackowski et al., 2009; 2011b). However, these imaging measurements provide only a static representation of bone tissue. Accordingly, in addition to static measure of bone tissue, it is also suggested that measures of a more dynamic nature are taken to better describe bone development during growth and maturation, which include blood biochemical markers of bone formation and resorption (Jurimae, 2010). One of the limitations using bone formation and resorption markers is that these markers represent an average turnover from all skeletal sites of the body and consequently are not site specific (Jurimae, 2010). However, it can be argued that to experience an increase in bone mineral values, elevations in bone formation markers would be necessary to overcome the increase in bone resorption markers (Jurimae, 2010). In addition, the knowledge of pubertal growth is necessary to correctly interpret the values of bone turnover, as the highest levels of bone turnover markers are observed at early puberty (Vaitkeviciute et al., 2016). It seems necessary to analyze the effect of childhood gymnastics athletic activity on bone mineral acquisition using both static and dynamic measures of bone development, as 60% of osteoporosis cases in adulthood are related to low bone mineral acquisition during childhood (Saggese et al., 2001).
To our knowledge, there is no overview of the specific effects of gymnastics activities on bone mineral accrual during growth and maturation in order to prevent possible osteoporosis in later adulthood. Therefore, we performed the present overview in order to assemble current evidence on this topic. Specifically, the aims of the current systematic review were to determine the differences in aBMD and BMC accrual between gymnasts and controls during the growing years and to describe different factors that could influence bone accrual.
Data sources and search strategy
This review followed the systematic review methodology proposed in the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) statement (Liberati et al., 2009). Studies were identified by searching within the electronic databases (Gomez-Bruton et al., 2013). The identification of studies was performed by searching within PubMed and SportDiscus (Gomez-Bruton et al., 2016).
The search was conducted up to September 2017. The first search was performed using the thesaurus provided by both databases, while the second search was performed with the following combination of terms: gymnasts and bone density (Gomez-Bruton et al., 2013; 2016). Two reviewers independently examined both databases to obtain the potential publications. Relevant articles were obtained in full, and assessed against inclusion and exclusion criteria described below. Inter-reviewer disagreements were resolved by consensus. Arbitration by a third reviewer was used for unresolved disagreements (Gomez-Bruton et al., 2013; 2016; Sioen et al., 2016).
The following inclusion criteria were used (Gomez-Bruton et al., 2013; 2016): 1) types of study designs: cross-sectional and longitudinal studies investigating the effects of gymnastics training programs on bone mass; 2) types of participants: children and adolescents; and 3) types of outcome measured: aBMD, BMC and bone area (BA) of the whole body (WB), lumbar spine (LS), femoral neck (FN) and forearm (FA) measured by DXA, and bone architecture of the tibia and radius measured by pQCT.
The following exclusion criteria were used (Gomez-Bruton et al., 2013; 2016): 1) studies without a control group that would permit comparison, 2) studies that measure BMD/aBMD but do not give specific WB, LS, FN or FA values, 3) studies that focus exclusively on bone metabolism markers without measuring bone with an imaging technique, 4) congress abstracts, dissertations and other similar unpublished data, 5) studies in languages other than English.
One author independently extracted bone density data from the included studies. The DXA-derived bone density data included in this review were obtained from the four frequently reported DXA scans: WB, LS, FN and FA. LS measures were taken from L1 to L4 or L2 to L4. Bone density values were generally extracted from the tables reported by the included papers in the systematic review. Estimates of volumetric bone density calculated by DXA (bone mineral apparent density), reported by some studies were not extracted. All included studies presented raw data.
Quality assessment of all studies included in this systematic review was done using the same quality assessment tool as Olmedillas et al. (2012), which grades articles on a scale of 7 points. This quality assessment tool has also been used in previous systematic reviews (Gomez-Bruton et al., 2013; 2016).
The initial search strategy identified 790 potentially relevant articles. Following the review of article titles and abstracts, and also excluding duplicate articles, where the same study results were described in more than one research article, the total number of articles was reduced to 75 potentially relevant papers for the...